Pharmacokinetics of antibiotics in natural and experimental superficial compartments in animals and humans.

Transcapillary exchange of antibiotics and other small molecules is diffusion driven and occurs in the capillary beds of the tissues. Small polar molecules are ionized at physiological pH and diffusion is pore-mediated. More lipophilic substances can also leave capillaries by the transcellular route. Splanchnic tissues have fenestrated capillary walls while somatic tissues have mainly micropores in their capillary walls. Under normal physiological conditions the ratio of capillary surface area to volume of fluid present (SA/V) is very large (> 100) and the rate of exchange of substances between capillaries, interstitial fluid and tissue fluid is extremely rapid. The structure of the interstitial space or ground substance, linking the capillary to the tissue cells, is designed to regulate the exchange of water, albumen and other solutes between the plasma and tissues. Interstitial space fluid (tissue fluid) is not simply an ultrafiltrate of plasma and has a specific chemical imbalance with plasma. Some antibiotics bind to serum albumen and it is claimed that this impairs their ability to penetrate into tissue fluids. However, the reduced concentration of albumen in the tissue fluid, relative to the plasma, is the main reason why percentage tissue penetration data based on total levels present are misleading and perpetuate the misconception that highly bound antibiotics (> 80%) have a reduced penetration potential. The majority of infections are localized in extracellular fluid. Several models have been developed to sample serially the extracellular compartment. They have yielded diverse concentration/time profiles even for the same antibiotic at similar sites (skin and subcutaneous tissues). It has been shown that the various profiles are a direct consequence of compartment SA/V, which can range from > 100 to < 10. In the preclinical situation, the blister model has proven popular and reproducible but it should be remembered that the delayed profile seen is an artefact of blister geometry (SA/V < 10). On the evidence available it is likely that bacterial infections, in the presence of acute inflammation, will enhance the rate of entry of agents, while the reverse is true in areas of chronic inflammation where pathological barriers are already in place.